Research Challenges in Wireless Networks of Biomedical Sensors

1
Research Challenges in Wireless Networks of
Biomedical Sensors

• Loren Schwiebert
• Wayne State University
• Department of Computer Science

Sandeep K. S. Gupta Arizona State
University Department of Computer Science and
Engineering
Jennifer Weinmann Wayne State University Departmen
t of Electrical and Computer Engineering
Additional Authors Ayad Salhieh, Vikram Shankar,
Valliappan Annamalai, Manish Kochhal, and Greg
Auner.
This material is based upon work supported by
the National Science Foundation under Grants
ANI-0086020 and DGE-9870720 and the Kresge Eye
Institute.
2
Typical Future Sensor Node

• Matchbox Size
• Battery Powered
• Power-conserving processor 100s of MHz
• 10s of MB of program and data memory
• Radio modem using TDMA
• Capable of running a scaled down version of Palm
  OS or Windows CE

3
Biomedical Smart Sensors??

• Smart Sensor Biomedical sensor/actuator with
  integrated circuitry
• Biomedical Implanted in the human body
• Form Small with limited power, encased in inert
  material
• Function Alleviate chronic diseases and
  disabilities, monitor health

4
Example Applications

• Glucose Level Monitor
• Transplant Organ Viability Monitor
• Blood Monitor
• Cancer Detection/Monitor
• Health Monitor
• Retinal and Cortical Prosthesis

5
Sensor-Based Visual Prostheses
Retinal Implant
Cortical Implant
6
Data Processing and Communication
7
Smart Sensor Retinal Interface
8
Multidisciplinary Research
Smart Sensors and Integrated Devices
Materials Characterization (Microstructure,
optical, electrical)
Materials Development
Materials Simulation, Device Simulation,
Design, and Testing

Device Development and Prototyping
Device Simulation Design and Testing
Materials Processing (Special lithography and
device fabrication development)

Electronic Integration Design
Data Communications and Interface Design
VLSI Circuit Development Intelligent system
Design and Development (Neuronet, logic)
Hybrid Technology and Packaging
Device Characterization, Testing, and Evaluation
9
Biomedical Sensor Constraints

• Limited Computation and Data Storage
• Ultra Low Power Consumption
• Wireless Communication
• Continuous Operation
• Inaccessibility

10
Biomedical Sensor Requirements

• Bio-Compatibility Material Constraints
• Robustness and Fault Tolerance
• Secure Data Communications
• Regulatory Requirements

Combination of Features Makes Biomedical Sensor
Networks Unique!
11
Research Approach

• Optimize across protocol layers
• Organize communication among sensor nodes
• Develop application-specific solutions
• Take advantage of biomedical sensor features
  fixed topology, pre-defined communication, and
  known membership
• Generalize these solutions

12
Communication Requirements of a Biosensor
Application

• Intra-sensor communication
• Data aggregation
• Distributed decision making
• Sensors to External controller (basestation)
  communication
• Downlink control operations.
• Uplink feedback.

13
Designing for Energy-Efficiency

• Sensor-sensor communication
• nearest-neighbor
• ad hoc
• Sensor-base station communication
• periodic
• long-distance sensor to base station
  communication
• Not energy-efficient to use the same routing
  protocol for both types of communication.

14
Research on Fixed Topologies

• Vary of Neighbors
• Trade-offs Exist
• Number of Hops
• Number of Receivers
• Amount of Contention
• Evaluate Power Usage
• Test Power-Aware Routing

15
Perf. Results Fixed Topologies

• Power-Aware Routing reduces Power Usage
• 3D is better than 2D
• 4 Neighbor Topology has lower Power Use
• Reason is always fewer Receptions

16
Cluster-Based and Tree-Based Approaches
17

Perf. Results Sensor-Base Station

• Cluster-based approach provides better
  energy-efficiency than the tree-based approach.
• True for a wide range of path loss exponents.
• For high path loss exponents, fewer clusters is
  better.

.
18
Current Research Emphasis

• Strict Power Management
• Efficient Wireless Spectrum Use
• Scalability Support as Many Sensing Elements as
  Possible
• Support Diagnostic Functionality
• Standardize Design with Other Research Groups

19
Wireless Networking is Key

• Novel Sensing Materials Exist
• Low-Power Electronics are Available
• Wireless Communication is Next Step
• Should Interoperate with Other Wireless Protocols
• Enormous Potential for Social Benefit

20
Retina and Cortical Implant Project Ophthalmology
Gary Abrams, MD Raymond Iezzi, MD Alexander
Dizoor, PhD Neurosurgery Pat McAllister,
PhD Robert Johnson, MD Janet Miller, B.S. Hun
Park, MD, PhD Todd Frances, M.S. Veterinarian Liz
Dawe, D.V.M. Arizona State University Sandeep
K.S. Gupta, PhD Valliappan Annalmalai, grad
student (NSF ITR) Karthik Jayaraman, grad student
(NSF ITR) Suresh Lalwani, grad student (NSF ITR)
Vikram Shankar, grad student (NSF ITR)

• Smart Sensors and Integrated Microsystems
• Gregory W. Auner, PhD
• Pepe Siy, PhD
• Loren Schwiebert, PhD
• Vaman Naik, PhD
• Ratna Naik, PhD
• Lowell Wenger, PhD
• Xiaoyan Han, PhD
• Yuriy Danylyuk, grad student
• Dan Durisin, engineer
• Francette Fey, grad student (NSF IGERT)
• Sam George, research assistant
• Changhe Huang, PhD
• Chantelle Hughes, grad student (NSF IGERT)
• Changli Jiao, grad student
• Manish Kochhal, grad student (NSF ITR)
• Michael Lukitsch, grad student (NSF IGERT)
• Marvie Nickola, grad student (NSF IGERT)
• Mona Safadi, grad student (NSF IGERT)